How do we do security in the world of AI and LLMs? A great place to start is with an OWASP project tasked with creating a standardized guideline for building secure AI applications with large language models such as ChatGPT. Enter OWASP Top Ten for LLMs, and Steve Wilson, the project leader.
You'll experience Large Language Models (LLMs) and their implications in AI. Steve explains how the introduction of ChatGPT marked a significant shift in the AI landscape. He elaborates on the concept of LLMs, their functioning, and the unique properties that emerge when used at a large scale.
Traditional OWASP Top Ten issues like SQL injection and broken authorization are still applicable when dealing with AI applications, and the OWASP API Top Ten could be layered onto these considerations. Think about it -- AI applications have web frontends.
A new discipline of AI security engineering is on the horizon, focusing on the security of large language models and the applications that access them. A focus on both AI safety AND security must occur.
We look forward to the release of the 1.0 version of the OWASP Top Ten for LLMs. Join the discussion today on OWASP Slack, and help form the new list.
FOLLOW OUR SOCIAL MEDIA:
➜Twitter: @AppSecPodcast
➜LinkedIn: The Application Security Podcast
➜YouTube: https://www.youtube.com/@ApplicationSecurityPodcast
Thanks for Listening!
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
